Image Data Compression and Noisy Channel Error Correction Using Deep Neural Network

Abstract

Everyday an enormous amount of information is stored, processed and transmitted digitally around the world. Neural Networks have been rapidly developed and researched as a solution to image processing tasks and channel error correction control. This paper presents a deep neural network (DNN) for gray image compression and a fault-tolerant transmission system with channel error-correction capabilities. First, a DNN implemented with the Levenberg-Marguardt learning algorithm is proposed for image compression. We demonstrate experimentally that our DNN not only provides better quality reconstructed images but also less computational capacity compared to DCT Zonal coding, DCT Threshold coding, Set Partitioning in Hierarchical Trees (SPIHT) and Gaussian Pyramid. Secondly, a DNN with improved channel error-correction rate is proposed. The experimental results indicate that our implemented network provides a superior error-correction ability by transmitting binary images over the noisy channel using Hamming and Repeat-Accumulate coding. Meanwhile, the network's storage requirement is 64 times less than the Hamming coding and 62 times less than the Repeat-Accumulate coding.